Recent studies from our laboratory and others suggest the existence of a complex cross- regulation of energy substrate metabolism between heart and periphery, mediated by peptides/proteins that add to insulin, glucagon or other classical regulators. Among the potential candidate mediators, particular attention deserve muscle-released, hormone-like factors collectively named myokines. One of the emerging members is follistatin-like protein 1 (Fstl1), which is produced also by the heart, as first showed by the Walsh laboratory, and can be therefore ascribed to the myokine subgroup named cardiokines. High circulating Fstl1 is a reliable marker of heart failure (HF), while, if acutely overexpressed, Fstl1 can protect the heart against ischemia-reperfusion injury. However, while the molecular biology and even the therapeutic potential of Fstl1 are being progressively elucidated and defined, virtually nothing is known yet about the integrative physiology and pathophysiology of this hormone. In particular, no studies have explored the potential role of Fstl1 as a metabolic modulator. Preliminary studies from Recchia and Walsh labs indicate that Fstl1 overexpression or infusion alters cardiac and systemic metabolism in mice and dogs. Therefore the present project will test the hypothesis that the relative fractions of Fstl1 produce by heart and skeletal muscle vary in response to physiological and pathological conditions and contribute proportionally to the autocrine and the reciprocal endocrine regulation of cardiac and systemic energy metabolism. Three specific aims will be pursued combining studies in dog and mouse models. Specific aim 1 will determine overall turnover and relative rate of Fstl1 production by cardiac and skeletal muscle at baseline and in response to exercise or to moderate and severe HF. These studies will be performed in chronically instrumented dogs under resting conditions and during exercise or the development of tachypacing-induced HF. Specific aim 2 will test whether Fstl1 modulates cardiac and systemic oxygen and energy substrate consumption under physiological conditions and in HF. These studies will be performed in mouse models of cardiac or skeletal muscle- selective Fstl1 loss of function and in chronically instrumented dogs. The underlying molecular changes and mechanisms will be determined in tissue samples, ex vivo, as well as in cultured cardiomyocytes and skeletal myofibers exposed to Fstl1. Specific aim 3 will test whether sustained, selective enhancement of cardiac Fstl1 synthesis can attenuate myocardial and/or systemic metabolic alterations and exert therapeutic effects in HF. These studies will be performed in cardio-specific Fstl1 knockout mice subjected to aortic constriction and in dogs with tachypacing-induced HF and subjected to cardiac gene transfer of Fstl1 carried by adeno- associated viral vectors. Combined studies in mice and dogs may lead to the translation of novel findings into a new biological therapy for HF.